Medtronic to distribute AI stroke detection and triage software

Dive Brief:

Medtronic will distribute clinical decision support software aimed at speeding stroke interventions as part of an agreement with Viz.ai, the company announced Monday. Financial terms of the deal were not disclosed.
Viz.ai is a Google Ventures-backed startup that won De Novo clearance early last year for its ContaCT application, which analyzes CT scans and flags suspected large vessel blockages via a smartphone notification and shares copies of the images to a neurovascular specialist within minutes. The startup received a 510(k) a few months later for an additional image processing software for CT perfusion scans. The company hasn't yet received marketing authorization abroad.
Medtronic said it aims to make Viz.ai's existing offerings available in every U.S. stroke center, building on the more than 200 hospitals that have already adopted the technologies.

Dive Insight:

Stroke is the fifth-leading cause of death in the U.S., and large vessel occlusions may be present in at least 30% of cases of acute ischemic stroke, according to a 10-year analysis recently published in the BMJ's Journal of NeuroInterventional Surgery. Medtronic already markets a stent that retrieves clots from occluded blood vessels to address such cases.

While FDA made clear in its February 2018 De Novo decision for ContaCT that the tool is only indicated for notification purposes, the agency said faster treatment could lessen the extent or progression of a stroke.

The FDA clearance was backed by a retrospective study of 300 CT images that compared Viz.ai's algorithm to neuroradiologists' expertise in detecting large vessel blockages in the brain. "Real-world evidence was used with a clinical study to demonstrate that the application could notify a neurovascular specialist sooner in cases where a blockage was suspected," FDA said after the De Novo announcement.

Viz.ai said at the time of the clearance the applications' median scan to notification time was less than six minutes, and, in more than 95% of cases, it said the software was faster than a specialist in deciding to notify a provider of suspected risk, saving 52 minutes on average.

Medtronic has already incorporated artificial intelligence in its diabetes unit. Its Sugar.IQ tool, used in conjunction with its continuous glucose monitor, leverages AI from IBM Watson Health to analyze how glucose levels react to factors like food intake and insulin dosing. In a similar vein, Medtronic last November acquired Nutrino, which uses AI to predict how nutrition may affect blood glucose.